| |
Common variation of IL28 affects gamma-GTP levels and inflammation of the liver in chronically infected hepatitis C virus patients
|
| |
| |
Download the PDF here
Jnl of Hepatology
published online 04 June 2010.
Hiromi Abe123, Hidenori Ochi123, Toshiro Maekawa12, C. Nelson Hayes123, Masataka Tsuge34, Daiki Miki123, Fukiko Mitsui123, Nobuhiko Hiraga123, Michio Imamura123, Shoichi Takahashi123, Waka Ohishi35, Koji Arihiro6, Michiaki Kubo7, Yusuke Nakamura78, Kazuaki Chayama123
"In this study, we focused on histological findings in the liver. We also analyzed viral and biochemical factors in patients chronically infected with HCV. We found that histological aspects of the liver (fibrosis and activity), HCV core amino acid substitutions, and gamma-GTP are associated with the polymorphism."
ABSTRACT
Background & Aims
A common genetic variation at the IL28 locus has been found to affect the response of peg-interferon and ribavirin combination therapy against chronic hepatitis C virus (HCV) infection. An allele associated with a favorable response (rs8099917 T), which is the major allele in the majority of Asian, American, and European populations, has also been found to be associated with spontaneous eradication of the virus.
Methods
As no studies have yet analyzed the effect of the polymorphism on biochemical and inflammatory changes in chronic infection, we analyzed a cohort of patients with chronic hepatitis C (n=364) for the effect of the IL28 polymorphism on viral, biochemical, and histological findings.
Results
We found that the proportion of HCV wild type core amino acids 70 and 91 was significantly greater (p=1.21x10-4 and 0.034) and levels of gamma-GTP significantly lower (p=0.001) in patients homozygous for the IL28 major allele. We also found that inflammation activity and fibrosis of the liver were significantly more severe in patients homozygous for the IL28 major allele (p=0.025 and 0.036, respectively). Although the higher gamma-GTP levels were also associated with higher inflammatory activity and fibrosis, multivariate analysis showed that only the IL28 allele polymorphism, sex, alcohol consumption, and liver fibrosis were independently associated with gamma-GTP levels (p=0.001, 0.0003, 0.0013, and 0.0348, respectively).
Conclusions
These results suggest that different cytokine profiles induced by the IL28 polymorphism resulted in different biochemical and inflammatory conditions during chronic HCV infection and contribute to the progression of liver diseases.
Introduction
Hepatitis C virus infection is one of the major causative agents of chronic hepatitis, liver cirrhosis, and hepatocellular carcinoma [1]. The best current therapeutic regimen is pegylated interferon and ribavirin combination therapy [2], [3]. Although the eradication rate of the virus has been improved by extending the treatment period from the standard 48 to 72weeks for genotype 1b infected patients, active viral replication still remains in nearly half of these patients [4].
Recent studies have identified both host and viral factors predictive of interferon therapy. Among the viral factors, a forty amino acid stretch in the NS5 region has been found to be predictive of response to interferon monotherapy [5], [6]. More recently, Akuta et al. identified amino acid substitutions in the core region (core aa70 and 91) that are predictive for the effect of interferon and ribavirin combination therapy [7], [8].
Among the host factors, many common polymorphisms in the human genome, including single nucleotide polymorphisms (SNP), have been identified [9], [10], [11], [12], [13]. We recently reported that a SNP in the MAPKAPK3 gene is associated with response to interferon therapy [14]. More recently, three groups of researchers found that several SNPs in the IL28 locus are related to the effectiveness of combination therapy [15], [16], [17]. We also performed a genome wide association study and confirmed that variation at the IL28 locus is related to the effectiveness of combination therapy (Chayama K, personal communication).
These viral and host factors must influence the natural course of viral infection. Host immune cells produce interferon and other cytokines in response to viral infection. For RNA viruses such as HCV, cellular sensors such as RIG-I detect the double stranded RNA and activate a pathway to produce cytokines, including alpha and beta interferons that trigger an antiviral response to eradicate the virus [18]. Genetic polymorphism of genes involved in innate immunity is likely to influence the strength and nature of this defense. In fact, a polymorphism in the IL28 locus has been reported to correlate with spontaneous eradication of HCV [19]. However, little is known about how these factors affect the course of chronic infection of the virus. In this study, we focused on histological findings in the liver. We also analyzed viral and biochemical factors in patients chronically infected with HCV. We found that histological aspects of the liver (fibrosis and activity), HCV core amino acid substitutions, and gamma-GTP are associated with the polymorphism.
Discussion
Polymorphism at the IL28 locus has been reported to be associated with the effectiveness of interferon and ribavirin combination therapy [15], [16], [17]. We have also found that the polymorphism is associated with the effect of interferon monotherapy on genotype 1b infected patients as well as genotype 2a infection in Japanese as well as Taiwanese patients (Chayama K, personal communication). The polymorphism has also been reported to be associated with spontaneous eradication of the hepatitis C virus [19]. As levels of IL28 gene transcripts have been reported to be higher in patients homozygous for the interferon response allele [16], [17], we hypothesized that the polymorphism is also associated with inflammation and progression of chronic hepatitis. As expected, there were significant associations between IL28 genotypes and histological inflammatory activity as well as the degree of fibrosis in chronically HCV infected patients (Table 5). It seems reasonable that the inflammation is stronger in patients with elevated IL28 production because this molecule induces expression of interferon stimulated genes, including some inflammatory cytokines. As the polymorphism is associated with the effect of interferon therapy, the interferon therapy performed before biopsy might alter the results. In fact, a part of patients in this study were treated with peg-interferon and ribavirin combination therapy and the treatment outcome was associated with IL28 genotypes and core amino acid substitutions (data not shown). However, when we analyzed the relation between the IL28 allele and histological findings or core amino acid substitutions in only treatment-na•ve patients, the results were unchanged, suggesting that the results obtained in this study are applicable without regard to history of interferon therapy.
Interestingly, the IL28 genotype was also associated with gamma-GTP levels and core amino acid substitutions, both of which are known to be predictive of response to interferon and ribavirin combination therapy [7], [8], [24]. The levels of liver enzymes such as ALT, AST, and gamma-GTP are usually higher in patients with high inflammatory activity. However, we observed that the levels were actually lower in patients with the favorable allele at the IL28 locus (the major allele in the Japanese population) (Figs. 1A and 2). A lower level of gamma-GTP has been reported to be associated with positive response to combination therapy. Further studies are needed to clarify the mechanism underlying the relation between gamma-GTP levels and therapy effectiveness. It would also be interesting to study the relationship between the IL28 allele and steatosis in the liver because gamma-GTP tends to be elevated in patients with steatosis, and steatosis caused by HCV core protein has been reported [25].
Similarly, viral wild type core amino acids 70 and 91 (i.e., core 70R and 91L), which were already known to be associated with positive response to combination therapy [7], [8], were also found to be associated with the favorable human IL28 alleles (Table 3). If viruses with wild type core amino acids 70 and 91 are more susceptible to interferon therapy, such strains should be less frequent in patients with higher cytokine levels. Viruses with wild type core 70 and 91 amino acids must therefore have some survival advantage in order to replicate in cells in which the level of IL28 production is high. Searching for a target molecule in the signaling cascade from sensing of the virus to production of IL28 might help resolve this question.
We also observed an association between high gamma-GTP levels and core amino acid 70 and 91 substitutions (Fig. 1C), although in multivariate analysis only IL28 genotype, liver fibrosis, sex, and alcohol consumption were significant predictors of gamma-GTP. It seems likely that these factors mutually interact in the presence of the virus and cytokines. Understanding these relationships will reveal the mechanism underlying the effective response to combination therapy and may suggest new strategies to cope with the hepatitis C virus.
Results
IL28 locus genotypes and viral and biochemical markers
We compared viral and biochemical markers with IL28 genotypes. First we analyzed the relationship between IL28 genotypes and substitutions in the HCV core protein amino acids 70 and 91, as well as the HCV genotype and the number of amino acid substitutions in the ISDR. As shown in Table 3, there are significant associations between amino acid substitutions in the core region and the genotype of the rs8099917 SNP at the IL28B locus. In particular, patients homozygous for the major IL28 allele were significantly associated with wild type core amino acid 70 (OR=0.30; p=1.21E-04). A similar trend is seen with core amino acid 91 substitutions (OR=0.50; p=0.034). Patients with more than one amino acid substitution in the ISDR region also tended to occur in patients homozygous for the major allele, although the difference was not statistically significant (Table 3). There was no correlation between the HCV genotype and the IL28 allele.
We further examined the relationship between IL28 and biochemical markers such as ALT, gamma-GTP, total cholesterol, HDL cholesterol, serum iron, and HCV RNA levels. Only the gamma-GTP level was significantly associated with the IL28 genotype. As shown in Fig. 1A, the gamma-GTP levels were lowest in the IL28 major allele homozygotes and highest in minor allele homozygotes. As drinking alcohol is known to elevate gamma-GTP levels, we examined the effect of alcohol intake in our cohort. As shown in Fig. 1B, there was an association between alcohol and gamma-GTP levels. As we found that the gamma-GTP level is higher in patients with core amino acid 70 substitutions (Fig. 1C), we performed multivariate analysis to examine what factors contribute to higher levels of gamma-GTP. As shown in Table 4, a simple regression analysis revealed that serum gamma-GTP levels were associated with sex, BMI, inflammation activity, liver fibrosis, alcohol consumption, and IL28 genotype, whereas in multiple regression analysis, sex, liver fibrosis, alcohol consumption, and IL28 genotype remained positively associated with serum gamma-GTP levels.
Histological findings and polymorphism in the IL28 locus
We then analyzed the relationship between the IL28 locus polymorphisms and histological findings. We divided patients into mild fibrosis (F0 and F1) and severe fibrosis (F2Ð4) as well as lower activity (A0 and A1) and higher activity (A2 and A3) and compared these factors against IL28 genotypes. As shown in Table 5, both inflammatory activity and fibrosis were significantly associated with IL28 genotype. Inflammation was more active (A2Ð3) in patients homozygous for IL28 major alleles (OR=1.75; p=0.025). Similarly, fibrosis was more severe in patients homozygous for IL28 major alleles (OR=1.66; p=0.036). We also performed analysis of the association of IL28 alleles and histological findings after adjusting for other factors that might influence the activity and fibrosis of the liver, such as age, gender, and alcoholic consumption. The IL28 allele was associated with F and A factors independently with adjustment for these predictive factors related to severity of liver fibrosis and inflammation (data not shown).
Relationship between histological activity, the IL28 allele, and gamma-GTP
As we described above, histological activity is more active in patients homozygous for IL28 major alleles. However, it seems contradictory that IL28 major allele homozygosity was associated with low levels of gamma-GTP, but severe activity was associated with high gamma-GTP. As shown in Fig. 2, however, when we compare the allele and activity the frequency of patients with higher activity (A2 and A3) were statistically more frequent in patients homozygous for major alleles (Fig. 2 and Table 5). When we compare gamma-GTP levels of A2 and A3 patients between patients homozygous for major alleles against the others, the levels were significantly lower in the major allele homozygous patients (Fig. 2). A0 and A1 patients showed similar results (Fig. 2). gamma-GTP levels are also significantly higher in patients with higher activities (A2 and A3) than in patients with lower activities (A0 and A1) (Fig. 2).
|
|
| |
| |
|
|
|
